Siglent Bench DMMs SDM3065X, SDM3055 and SDM3045X

[KungFuJosh]

But this was reported on a SDM3065X. What is there to hack on that model? The only thing I could think of is adding the scanner card, but that requires hardware mods -- I think the firmware already has support for it.

🤷I dunno. it could just be a hardware revision change with the same firmware. Point being it's not something new, and it's definitely not a bug.

I know from the 3045X improvement thread that it prevents the upgrade from happening for now.

[Brad808]

I just bought an SDM3055 and I have a basic question for you guys. Is there no way to get a single beep in diode mode? I read through the manual and on page 33/34 it looks like it should be able to, but I can't get it to operate that way unless I'm not understanding it correctly. I'd like it to be like my fluke 289 = nothing on open, single beep for working diode, and continuous for short. Is that possible? Thanks

[Veteran68]

I just bought an SDM3055 and I have a basic question for you guys. Is there no way to get a single beep in diode mode? I read through the manual and on page 33/34 it looks like it should be able to, but I can't get it to operate that way unless I'm not understanding it correctly. I'd like it to be like my fluke 289 = nothing on open, single beep for working diode, and continuous for short. Is that possible? Thanks

Not sure about the SDM3055 but on my SDM3065X you get a continuous diode beep when voltage drop is below the set threshold. There is no distinction between working diode and a short.

When in diode mode, the 3065X has 3 options along the bottom: Beeper On/Off, Threshold voltage, and Volume. Do you not see those options on the 3055?

Honestly in my experience, and I have around 30 DMMs, is that very few follow Fluke's pattern of single vs continuous beep. Not even all Fluke's do it (my 87V and 117 do, my 17B Max and 101 do not beep at all). My Agilent 34401A does not beep at all in diode mode, whether diode junction or shorted, nor do any of my Brymen meters. Surprisingly, the Uni-T 61E+ does have the Fluke-style single vs continuous diode beep.

[Brad808]

Not sure about the SDM3055 but on my SDM3065X you get a continuous diode beep when voltage drop is below the set threshold. There is no distinction between working diode and a short.

When in diode mode, the 3065X has 3 options along the bottom: Beeper On/Off, Threshold voltage, and Volume. Do you not see those options on the 3055?

I do see those options and tried playing with the threshold, but it seems it will only go continuous or nothing. I must be misreading this line from the manual somehow, because it sounds like exactly what I want, but evidently it's not. Ah well, not the end of the world, just a feature I find handy on my Fluke 289.

[skander36]

Also KS 34465A does single beep on diode test mode. It's nice but not the end of the world if it wasn't.

[Veteran68]

Not sure about the SDM3055 but on my SDM3065X you get a continuous diode beep when voltage drop is below the set threshold. There is no distinction between working diode and a short.

When in diode mode, the 3065X has 3 options along the bottom: Beeper On/Off, Threshold voltage, and Volume. Do you not see those options on the 3055?

I do see those options and tried playing with the threshold, but it seems it will only go continuous or nothing. I must be misreading this line from the manual somehow, because it sounds like exactly what I want, but evidently it's not. Ah well, not the end of the world, just a feature I find handy on my Fluke 289.

(Attachment Link)

Interesting, the 3065X manual doesn't mention a single beep at all. Just a (continuous) beep when below the threshold voltage. I wonder if they used to support it but dropped it in later firmware. I'd prefer that it did support the single beep and am curious why they'd remove support for it.

[Veteran68]

My Agilent 34401A does not beep at all in diode mode, whether diode junction or shorted, nor do any of my Brymen meters.

Correction: my eevBlog Brymen BM786 does mimic the Fluke behavior of single beep on diode junction and continuous beep on short. Nice! Further demonstrating that the BM78x is one of the best electronics handheld DMMs you can get.

However neither the flagship BM869s nor the FLIR DM64 (IIRC a BM235 inside) beep at all, even for a short.

[MathWizard]

Today using continuity mode on my SDM3065X, I noticed it was taking nearly 1 second for it to beep with shorted probes, limit is set to 2ohms. They are pretty clean and it was normal speed early, under a few hundred miliseconds.

I rebooted and it was still slow beeping, then I used resistance mode, and when I went back to continuity, it was back to normal.


Anyone have that happen before ? I don't think I've had it before. The DMM was on for 10-20min before I used it, and already on Cont.mode.

[ComradeXavier]

Today using continuity mode on my SDM3065X, I noticed it was taking nearly 1 second for it to beep with shorted probes, limit is set to 2ohms. They are pretty clean and it was normal speed early, under a few hundred miliseconds.

I rebooted and it was still slow beeping, then I used resistance mode, and when I went back to continuity, it was back to normal.

Anyone have that happen before ? I don't think I've had it before. The DMM was on for 10-20min before I used it, and already on Cont.mode.

I haven't seen that problem specifically, but you might want to install the 3.01.01.12R1 firmware update if you haven't already. They significantly improved the continuity mode performance. Before that I would have said that 500ms latency was typical.

[MathWizard]

Toady I used the DMM on Cont. and turned it off, then now 2-3hrs later when I turned it back on, it was slow again. But changing to Ohms and back to Cont. and it's back to normal. I use the setting where it remembers your settings (I wish I kenw about that in the 1st 1-2yrs I had it and my Siglent AWG)

I updated the firmware this spring/summer, that was easy so I'll check again.

[KungFuJosh]

I don't have a 3065, but I've never had that issue with either of the other 2 models.

[klausES]

...But changing to Ohms and back to Cont. and it's back to normal...

If this is repeatable for you, it indicates a bug.
A similar behavior and reset the error by the fact that you go briefly to another measurement type and back again
(only in another context by the 3055) there was a bug in a previous firmware at the 3055.

[KungFuJosh]

I have an accuracy question. The 3055 datasheet for AC voltage says: 200 mV, 45 Hz – 20 kHz, 0.2 + 0.05

I'm testing 200mV at 1kHz. On my scope I get 200.04mV RMS.

SDM3055X-E says 202.36mV, and SDM3055 says 202.21mV. Doesn't that seem high? Shouldn't the above margin put it at 200.45mV? Or is that 0.2 a typo, and it should really be 1.2%?

Room temp is 20C, which is within the range for the above specs.

[Martin72]

Hi,

On my scope I get 200.04mV RMS.

The scope is not to be trusted in terms of accuracy, I would ignore it.
I would trust the SDM3055 meters, unless you know that the signal is exactly 200mV...
Source for the 200mV signal ?

[KungFuJosh]

Source for the 200mV signal ?

I'm testing the signal level on the ST2832 LCR.

Why not trust the scope? I can step through by 10mV from 10mV to 600mV, and the scope is dead on, with maybe a small µV variance. Then after 700mV to 1V there's a small mV difference, maybe +9mV or so. The rises a little higher over 1V, but it's consistent and repeatable for whatever that's worth.

[Martin72]

I'm testing the signal level on the ST2832 LCR

In normal mode the accuracy of the output level is:

± (10%×preset value+2mV)

I interpret this to mean that the signal can be anywhere between 182mV....222mV.
Your SDM3055 meters measure 202mV, your scope 200mV.
Now compare the vertical accuracy of your scope with the accuracy of your meters...

[KungFuJosh]

I'm testing the signal level on the ST2832 LCR

In normal mode the accuracy of the output level is:

± (10%×preset value+2mV)

I interpret this to mean that the signal can be anywhere between 182mV....222mV.
Your SDM3055 meters measure 202mV, your scope 200mV.
Now compare the vertical accuracy of your scope with the accuracy of your meters...

The funny thing is that the DMMs spec sheet say 200mV to 750VAC. After letting the meters run for a week, they're all (scope + DMMs) basically in agreement for 20mV being ~20.2mV.

Before letting them run that long, the X-E said 18mV, and the 3055 said 27mV. So I'll call that an improvement.

[Martin72]

Which meters do you use ?

3055, what is "X-E" ?

Scope is the SDS2504X plus, right ?

Edit:

Regarding your post edit with the pictures:

Use more vertical space when measuring.
I would not measure the 200mV signal at a setting of 500mV/div.

[KungFuJosh]

Which meters do you use ?

3055, what is "X-E" ?

Scope is the SDS2504X plus, right ?

SDM3055 and SDM3055X-E (improved SDM3045X with China firmware).

Regarding your post edit with the pictures:

Use more vertical space when measuring.
I would not measure the 200mV signal at a setting of 500mV/div.

Sure, but I was ignoring the divs and just reading the measured RMS value.

[KungFuJosh]

J-R gave me the excellent idea to use my AWG as a voltage ref. Looks good, even at lower mV.

[Kleinstein]

Why not trust the scope?
[/quote]

The specs for scopes are usually for something like +-1% to +-3% for the amplitude accuracy. So for the absolute scale the scope is rather poor.
One can however use the scope to check the frequency response to a certain degree. A DSO should have a resonable flat response between 1 kHz and 1 MHz.
Even there one has to be somewhat careful as there are compensated resistive / capacitive divders that can lead to some change in the amplitude when going from the resistive to capacitive dominated band.

[KungFuJosh]

The specs for scopes are usually for something like +-1% to +-3% for the amplitude accuracy. So for the absolute scale the scope is rather poor.
One can however use the scope to check the frequency response to a certain degree. A DSO should have a resonable flat response between 1 kHz and 1 MHz.
Even there one has to be somewhat careful as there are compensated resistive / capacitive divders that can lead to some change in the amplitude when going from the resistive to capacitive dominated band.

Correct me if I'm wrong, but I believe the spec you're referring to is DC Accuracy specifically, not amplitude accuracy. I'm measuring AC voltage with AC coupling.

In any case, 3% isn't terrible in the mV range. The datasheet for the LCR I'm measuring states an accuracy of 10% for test signal level. I believe that makes the scope worthy in this use case. 😉

More importantly, I'm a "trust but verify" kinda guy. If the measurements are dead on, verifiable, and repeatable...

[Martin72]

In any case, 3% isn't terrible in the mV range. The datasheet for the LCR I'm measuring states an accuracy of 10% for test signal level. I believe that makes the scope worthy in this use case. 😉

A few had answered your original question:
https://www.eevblog.com/forum/testgear/siglent-new-bench-dmm-sdm3055/msg5304295/#msg5304295

Now 3% is not so bad...
But it doesn't change the fact that a scope measures "less accurately" than a 5.5 digit meter.
We had a colleague at work who measured a coil and swapped it with the LCR meters available in the house, whether calibrated or not, until the result met his expectations.

[KungFuJosh]

Now 3% is not so bad...
But it doesn't change the fact that a scope measures "less accurately" than a 5.5 digit meter.

Of course a DMM is expected be more accurate. But when we're talking about an acceptable margin of +-10%, 3% is plenty accurate enough (despite it being unclear of that's the correct number).

The claimed 0.2% accuracy of the DMM is nice, but not essential in this case. Using the AWG as a ref and seeing them agree is certainly helpful.

We had a colleague at work who measured a coil and swapped it with the LCR meters available in the house, whether calibrated or not, until the result met his expectations.

🤣 Next thing you know they'll be p-hacking data for marketing purposes.

[H.O]

I'd certainly trust the meters over scope in this case.
The scope has an 8bit ADC.
With 8 vertical divisions on the screen and 500mV per division (as in the screenshot with the 200mV RMS signal) the full scale input to the ADC will be 4V (overrange not accounted for).
At 4V full scale, each "count" out of the converter "is worth" 15.625mV. So ~16mV is the resolution at that vertical scale.

Displaying a 600mV p-p signal on a 4000mV p-p screen effectively turns your 8bit ADC into a 5bit (and change) ADC.